UPLC-ESI-QTOF-MS Profiling of Phenolic Compounds from Eriocephalus africanus: In Vitro Antioxidant, Antidiabetic, and Anti-Inflammatory Potentials

The present study investigated phenolic compounds, antioxidant, antidiabetic, and the anti-inflammatory potentials of methanolic and chloroform extracts of Eriocephalus africanus. The methanolic extract included, polyphenols (112 ± 2.81 mg gallic acid equivalent (GAE)/g), flavonols (76.12 ± 7.95 mg quercetin equivalents (QE)/g); antioxidant capacity (Ferric Reducing Antioxidant Power (FRAP) (752.64 ± 89.0 μmol of ascorbic acid equivalents (AAE) per g dry weight (µmol AAE/g), 2,2-dyphenyl-1-picrylhydrazyl (DPPH) (812.18 ± 51.12 Trolox equivalents per gram of dry mass of plant extracts (μmol TE/g), TEAC (631.63 ± 17.42 µmol TE/g)), while the chloroform extract included polyphenols (39.93 ± 1.36 mg GAE/g), flavonols (44.81 ± 3.74 mg QE/g); antioxidant capacity, DPPH (58.70 ± 5.18 µmol TE/g), TEAC (118.63 ± 3.74 µmol TE/g) and FRAP (107.10 ± 2.41 µmol AAE/g). The phytochemicals profiling performed by UPLC-ESI-QTOF-MS revealed some important polyphenols, predominantly flavonoids, that could be responsible for the antioxidant capacity and biological effects. Both extracts demonstrated a dose-dependent manner of the alpha-glucosidase inhibition with an IC50 between 125 and 250 μg/mL for methanolic extract, while the chloroform extract was at 250 μg/mL. In the L6 myoblasts and C3A hepatocytes, the methanolic extract slightly increased the utilization of glucose, and both extracts exhibited a dose-dependent increase in the glucose uptake in both cell types without significantly increasing the cytotoxicity. Furthermore, both extracts exhibited an anti-inflammatory potential and the findings from the present study could serve as a baseline for further research in the development of pharmaceutical agents.


Introduction
The number of people with metabolic diseases, such as diabetes and obesity, is increasing across the globe [1]. According to a 2017 estimate from the International Diabetes Federation, 451 million individuals globally have diabetes, and by 2045, there will probably be 693 million diabetic cases [2]. The excessive consumption of carbohydrates and triglycerides is one of the main factors that contribute to the metabolic syndrome [3]. It is relatively common and affects people all over the world who consume too much energy and do not exercise regularly. Type 2 diabetes (T2D), which is one of these metabolic syndromes, and is defined by hyperglycaemia and the impaired carbohydrate metabolism, it is a substantial economic burden and a leading cause of morbidity and death, globally [4,5]. The majority (>95%) of newly diagnosed diabetics have type 2 diabetes (T2D), which is driven by the pancreatic beta cell failure and insulin resistance [6]. Obesity is the cause of insulin resistance and pancreatic beta-cell death, which establishes the connection between T2D and obesity [7]. The onset and progression of T2D complications are significantly influenced by postprandial blood glucose levels [8]. There is a connection between diabetes

Phytochemical Compounds and Antioxidant Activity
Diabetes mellitus is one of the most prevalent endocrine diseases and poses a serious threat to global public health. The search for a safer option has become imperative because the conventional drugs used to manage diabetes are associated with side effects. Remarkably, natural antioxidant-rich products containing phenolic compounds have gained attention for the treatment of oxidative stress-related diseases. The most prevalent secondary metabolites identified in plants and widely spread throughout the plant kingdom are phenolics [32]. Plant secondary metabolites serve many functions, including those related to innate immunity [33], defense response signaling [34], and plant growth and development processes [35]. Interestingly, the potential health advantages of plant polyphenolics have been reported in various studies [36][37][38]. Plant phenolics have antioxidant properties that have been scientifically demonstrated to prevent a variety of chronic diseases and oxidative stress-related illnesses [39,40]. In the present study, both polar (70% methanol) and non-polar solvents (chloroform) were used as the extraction solvents. From 100 g of the plant powder extracted with 1 L of the extraction solvents, 22.4 g and 15 g yields were obtained from the methanol and chloroform extracts, respectively.
The polyphenol and flavonol contents in the crude methanolic and chloroform extracts of E. africanus were investigated and the results are represented in Table 1. It was observed that the polyphenol content (112 ± 2.81 mg GAE/g) in the methanolic extract of E. africanus was higher, as compared to the chloroform extract (39.93 ± 1.36 mg GAE/g). Similarly, the flavonol content of the methanolic extract of E. africanus was higher (76.12 ± 7.95 mg QE/g), when compared with that of the chloroform extract (44.81 ± 3.74 mg QE/g). Concomitantly, the antioxidant capacity of the extracts was also investigated, and it was recorded that the methanolic extract has a higher antioxidant capacity (FRAP-752.64 ± 89.0 µmol AAE/g; DPPH-812.18 ± 51.12 µmol TE/g; TEAC-631.63 ± 17.42 µmol TE/g), as compared to the chloroform extract (FRAP-107.10 ± 2.41 µmol AAE/g; DPPH-58.70 ± 5.18 µmol TE/g; TEAC-118.63 ± 3.74 µmol TE/g). The higher antioxidant capacity of the methanolic extract, as compared to the chloroform extract could be due to the high contents of polyphenols and flavonols. The higher amount of both polyphenols and flavonols in the methanolic extract, compared to the chloroform extract, is an indication of the possible therapeutic role of plants and plant-based products [32,41]. polar and polar phenolic substances, such as phenolic acids and flavonoid glycosides, are frequently extracted using methanol [42]. The extraction solvent has a significant effect on the nature of the phytochemical compounds that will be extracted. The solvent polarity influences the extraction of the phenolic compounds from medicinal plants, and this could be responsible for the higher phenolic compounds and antioxidant capacity of the methanolic extract of E. africanus over the chloroform extract [32]. The higher antioxidant potential of the methanolic extract is mostly due to the nature of the phenolic compounds [43,44].

Phytochemical Analysis of the Methanol and Chloroform Extracts of E. africanus
Sixty-two compounds were identified in E. africanus methanol and chloroform extracts, through a high-resolution UHPLC-QTOF-MS analysis (Tables 2 and 3, Figures 1 and 2). The compounds were identified using their full mass spectra and MS/MS fragmentation patterns, and then compared to the published data or MS/MS databases, such as Metlin, MassBank, PubChem, etc. In addition, a standard mixture (Figure 3) was used to completely identify metabolites, using their retention time and fragmentation. The standard mixture was prepared by injecting 50 ppm of each of the chemical marker compounds. A total of 48 metabolites were identified from the methanol extract, compared to 28 metabolites identified from the chloroform extract of E. africanus.

Analysis of the Phenolic Compounds and Other Metabolites
In total, 62 metabolites could be identified from the methanol and chloroform extracts of E. africanus. Flavonoids were the most identified compounds among which flavones and flavonols were the most represented. Other classes of organic compounds included carboxylic acids, hydroxicinnamic acid, flavan-3-ols, and coumaric and furanocoumaric compounds.

Analysis of the Carboxylic Acids
Two carboxylic acid peaks could be identified, i.e., peaks 7 and 36. The fragmentation of the carboxylic acids occurred through the release of one or two water (18 Da

Analysis of the Hydroxycinnamic Acids
In the present study, nine hydroxycinnamic acids or derivatives were tentatively or completely identified (peaks 1, 2, 3, 4, 8, 12, 13, 15, and 16). Six quinic acid (QA) derivatives were identified, including two earlier reported: caffeoyl quinic acid (CQA), from this plant (peaks 3 and 4) [27], and di-CQAs (12, 13, 15 and 16). Their spectra showed a deprotonated molecular ion at m/z 353 of mono-CQAs and m/z 515 of di-CQA. The fragmentation in the MS/MS produced m/z 191 (QA), which gave a dehydrated quinic acid moiety (m/z 173) and caffeic acid (m/z 179), as a prominent fragment. The 3-O-CQA (neochlorogenic acid) that eluted at RT 7.11 min and 1-O-CQA (chlorogenic acid) (UV maxima at 326 nm) were differentiated by the intensity of the characteristic ions of the chlorogenic acids and the retention time comparison with that of the standard. In the former, m/z 179 is the base peak and in the latter, m/z 191 is the base peak [46,47]. For di-O-CQAs, the more the additional caffeoyl groups attached to the less free equatorial hydroxyl groups (owing to steric interactions) retained in the quinic acid residue, the stronger the retention [48]. That is, the loss of the caffeoyl group (C) is likely to be in the order; [46,48]. This enabled peaks 12, 13, 15, and 16 to be identified as 1,5 di-O-CQA, 3,4 di-O-CQA, 3,5 di-O-CQA and 1,4 di-O-CQA, respectively, since the elution order is 1,5-diCQA << 3,4-diCQA << 3,5-diCQA << 1,4-diCQA [46]. The prominent ions for 3,5-di-CQA have m/z 179 as base peaks, consistent with earlier studies [47,49,50]. E. africanus has also earlier been reported to be one of the sources of 3,4 di-O-CQA, 3,5 di-O-CQA, and 1,4 di-O-CQA [27]. Peak  identified in peak 1 as p-coumaric acid ethyl ester, having shown a deprotonated ion at m/z 191 with the MS 2 characteristic ion of the coumaroyl at m/z 119 [coumaric acid-CO 2 ] -. An acylated derivative of caffeic acid, a caffeoyl-hexuronide derivative was identified with adduct [M − H + Cl] -, m/z 554 with fragments m/z 355 (after the loss of 162 Da, probably that of the caffeoyl moiety or of a hexose) and then of a hexuronide acid (m/z 113), as earlier reported from this plant [27].               The retro-Diels-Alder was used to study the flavonoid fragmentation pathways (RDA) and 14 flavones were identified. Peaks 6, 27, 32, and 33 were identified to belong to that apigenin or its C-linked sugar conjugate, as earlier reported [45], or newly reported from this plant using the fragmentation pattern in earlier reports. Peaks 14,21,22,27,28,29,30,31,32,33,51,55 m/z 287 and the MS 2 main fragments 151 [ 1,3 A -], formed through the retrocyclization RDA cleavages of the C-ring of the aglycone involving 1 and 3 bonds (bonds 1 and 3 refer to the O-C-2 and C-3-C-4 bonds of the C-ring) [51], which is consistent with that of the aglycone eriodictyol. The conjugation with hexuronide (176 Da), formed an ion at [M − H] -; m/z 463 and thus was tentatively identified as eriodictyol-O-hexuronide for peaks 9 and 16. Identification of eriodictyol and eriodictyol-O-hexuronide in E. africanus, is consistent with an earlier report [28]. The compound in peak 26 was identified as naringenin, due to

UPLC-QTOF-MS Quantitation of the Phenolic Compounds
To improve on the impact of the study, the phenolic compounds were quantified using marker compounds, as described earlier [21]. The UV/vis absorptions, capable of distinguishing the phenolic subclasses, were considered a starting point for compound quantification. Based on the qualitative analysis, four chemical markers, namely phenolic acids and derivatives (neochlorogenic acid (peak 1), caffeic acid (peak 3), ferulic acid (peak 6), and coumaric acid (peak 5)), flavonol (peak 7), dihydrochalcone (peak 8), and flavano-3ols (peak 2/4), were selected for the simultaneous quantitative determination ( Figure 4). They were the most abundant in many reported plant methanol and chloroform extracts. They have been well represented in this study in the methanol and chloroform extracts of E. africanus. Based on the UV spectrum of the markers, the UV detection wavelengths were chosen. The phenolic acids and the derivatives had the strongest UV absorption at 300, 308, and 325 nm, whereas flavonols and flavanones at 254, 255, and 354 nm, flavones at 284 and 350 nm, and finally flavan-3-ol at 277 nm. The limits of quantification and detection, LOQ and LOD, respectively, presented in Table 4 were calculated by the parameters of the analytical curves (standard deviation of the response and slope) in our previous studies [21]. The standard deviation of the y-intercepts of the regression lines was used as the standard deviation of the blank. The LODs and LOQs were estimated as 3.3 and 10 times the standard deviation of the blank/slope ratio of the calibration curve, respectively. According to the LODs and LOQs, the phenolic compounds were highly detectable and quantifiable using the methods specified in the 70% ethanol extract of E. africanus ( Figure 4). (Table 4). This approach may be used to profile the phenolic compounds from the samples examined, according to the observed limits of detection and quantification. Neochlorogenic acid, caffeic acid, ferulic acid, coumaric acid, rutin, catechin, epicatechin, and phloridzin equivalents have been used to express the quantity of the phenolic constituents like in the previous studies [21,53].

Alpha-Glucosidase Inhibition
Natural plant products are well-known for their multipurpose uses as preventative and therapeutic agents, as well as their low toxicity and adverse effects [54]. The adverse effects of diabetes drugs necessitated the development of innovative approaches for DM treatment [55]. Multiple studies have shown that plant extracts can function as α-glucosidase inhibitors, implying that they can be used to alleviate hyperglycemia [55][56][57]. The alpha-glucosidase inhibition of the crude methanolic and chloroform extracts of E. africanus was investigated and the results are depicted in Figure 5. It was observed that both extracts exhibited dose-dependent alpha-glucosidase inhibitory activities. Each of them displayed a similar inhibitory ability, with the sample methanolic extract exhibiting a higher inhibition of α-glucosidase (above 70%) at its highest tested concentration of 1000 μg/mL. The IC50 of these extracts are estimated as follows: the methanolic extract between 125 and 250 μg/mL, while the chloroform extract is approximately 250 μg/mL. A dosedependent manner of the α-glucosidase inhibition has been reported by other researchers on some plant extracts in the literature [58]. Previous studies have shown that plant extracts that exhibited excellent antioxidant activities are more likely to demonstrate an αglucosidase inhibition, depending on the type of active phytochemical compounds, which can vary from species to species, and rely on extraction procedures [59,60].

Alpha-Glucosidase Inhibition
Natural plant products are well-known for their multipurpose uses as preventative and therapeutic agents, as well as their low toxicity and adverse effects [54]. The adverse effects of diabetes drugs necessitated the development of innovative approaches for DM treatment [55]. Multiple studies have shown that plant extracts can function as α-glucosidase inhibitors, implying that they can be used to alleviate hyperglycemia [55][56][57]. The alphaglucosidase inhibition of the crude methanolic and chloroform extracts of E. africanus was investigated and the results are depicted in Figure 5. It was observed that both extracts exhibited dose-dependent alpha-glucosidase inhibitory activities. Each of them displayed a similar inhibitory ability, with the sample methanolic extract exhibiting a higher inhibition of α-glucosidase (above 70%) at its highest tested concentration of 1000 µg/mL. The IC 50 of these extracts are estimated as follows: the methanolic extract between 125 and 250 µg/mL, while the chloroform extract is approximately 250 µg/mL. A dose-dependent manner of the α-glucosidase inhibition has been reported by other researchers on some plant extracts in the literature [58]. Previous studies have shown that plant extracts that exhibited excellent antioxidant activities are more likely to demonstrate an α-glucosidase inhibition, depending on the type of active phytochemical compounds, which can vary from species to species, and rely on extraction procedures [59,60]. The results of this study are consistent with the findings of Bhatia et al. [54], who found a methanolic extract of Cornus capitata that displayed an inhibitory activity with a 98.37% inhibition (IC50 12.5 μg/mL). In nature, the phenolics and flavonoids have the capacity to neutralize the free radicals (ROS) and can block the α-glucosidase to regulate blood sugar levels [61]. As previously documented, the antidiabetic effects of plant extracts were most likely related to the polyphenols found in plant extracts, and the inhibition of α-glucosidase was identified as the key target for the treatment of high postprandial blood glucose levels [62,63]. The final stage of starch digestion is catalyzed by α-glucosidase, which hydrolyzes the terminal glucose molecules from the non-reducing ends of oligosaccharides. α-glucosidase, also known as maltase-glucoamylase (MGAM), and The results of this study are consistent with the findings of Bhatia et al. [54], who found a methanolic extract of Cornus capitata that displayed an inhibitory activity with a 98.37% inhibition (IC 50 12.5 µg/mL). In nature, the phenolics and flavonoids have the capacity to neutralize the free radicals (ROS) and can block the α-glucosidase to regulate blood sugar levels [61]. As previously documented, the antidiabetic effects of plant extracts were most likely related to the polyphenols found in plant extracts, and the inhibition of α-glucosidase was identified as the key target for the treatment of high postprandial blood glucose levels [62,63]. The final stage of starch digestion is catalyzed by α-glucosidase, which hydrolyzes the terminal glucose molecules from the non-reducing ends of oligosaccharides. α-glucosidase, also known as maltase-glucoamylase (MGAM), and sucrose-isomaltase (SI), which is a membrane-bound enzyme found near the brush border of the epithelial cells of the small intestine [64,65].
Alpha-glucosidase inhibitors slow the rate at which glucose may be absorbed and distributed in the body, by slowing down the digestion of carbohydrates and this prevents postprandial hyperglycaemia. For α-glucosidase inhibitors to successfully prevent the hydrolysis of oligosaccharides, they must ideally bind to all four catalytic domains of the enzyme. Acarbose is one such inhibitor that functions via a competitive inhibition mechanism [66]. As a result, α-glucosidase inhibitors are widely used as oral antidiabetic drugs in the early stages of T2D, to treat obesity and postprandial hyperglycemia. Several studies have reported that flavonoids have potential inhibitory effects on α-glucosidase [67][68][69]. Furthermore, it has been established that α-glucosidase inhibitors have a protective influence on the blood vessels by lowering the postprandial glucose levels, which are linked to endothelial dysfunction, cardiovascular disease, and stroke [70].

Glucose Uptake
Medicinal plants are used in the treatment of diabetes, and this has generated attention, due to their efficiency and cost-effectiveness [56]. In the search for plant-based products for managing diabetes, the effect of crude methanolic and chloroform extracts of E. africanus was investigated on the glucose uptake and utilization in C3A hepatocytes and L6 myoblasts. Following the 24 h chronic exposure to the samples, the glucose uptake was determined over a 4 h period, as a function of the amount of glucose in the spent culture medium. The positive control, insulin, was added just prior to the experiment to represent an acute effect. The glucose utilization was measured after 24 h of treatment and reflects the cumulative change in the concentration of the glucose during treatment. Considering that glucose absorption through the glucose transporter proteins is the rate-limiting step for the cellular glucose uptake, it may be assumed that the amount of glucose left over indirectly reflects the glucose uptake, even when the intracellular glucose concentration is not known.
The effect of the extracts on the glucose utilization and the uptake was determined in C3A and L6 cells ( Figure 6). Both extracts demonstrated a dose-dependent increase in glucose uptake in both cell lines tested. The methanolic extract of E. africanus exhibited an increased glucose utilization in L6 myoblasts at all tested concentrations (25-100 µg/mL), whereas chloroform extract of E. africanus showed no significant effects. In the C3A hepatocytes, the methanolic extract showed an increase in the glucose utilization at a treatment concentration of 100 µg/mL, while the chloroform extract was at 25-50 µg/mL. Moreover, the methanolic extract increased the glucose uptake in the L6 myoblasts at all tested concentrations (25-100 µg/mL), while the chloroform extract displaced the glucose uptake from 50-100 µg/mL. No significant cytotoxic effects were observed for all test samples in the L6 cells or C3A cells (Figure 7). The main site for utilizing the glucose following meals is skeletal muscle, which also plays a significant role in maintaining the glucose homeostasis [81]. Numerous compounds produced by plants have been attributed to the positive impact on the glucose transport and metabolism in the skeletal muscle cells. This study demonstrated that treatment with a plant extract boosted the glucose absorption in a concentration-dependent manner, indicating that the higher the drug concentration, the stronger the effect [82]. Considering that the adipocytes are the primary site of the insulin action, they are crucial for both the control of the whole-body glucose homeostasis and glucose metabolism [83]. It is important to mention that the biological effect exhibited by these extracts could be due to the synergistic effect of the bioactive compounds identified in them. This study demonstrated that treatment with a plant extract boosted the glucose absorption in a concentration-dependent manner, indicating that the higher the drug concentration, the stronger the effect [82]. Considering that the adipocytes are the primary site of the insulin action, they are crucial for both the control of the whole-body glucose homeostasis and glucose metabolism [83]. It is important to mention that the biological effect exhibited by these extracts could be due to the synergistic effect of the bioactive compounds identified in them.

Pancreatic β-Cell Proliferation
The beta cell apoptosis is a hallmark of both type 1 and types 2 diabetes [84,85]. The islets of the pancreas' beta cells undergo a selective apoptosis, which results in an insulin insufficiency and persistent hyperglycaemia in type 1 diabetes (T1DM). In type 2 diabetes (T2DM), the functional deficiencies and reduced beta cell mass both lead to a beta cell failure [86]. Once T2DM is present, abnormal amounts of metabolic factors cause the beta cell death, which is followed by apoptosis [87]. Strong activators of the β-cell proliferation can restore the pancreatic function through neogenesis [88]. Therefore, showing the therapeutic potential in the relief and or reversal of both the T1DM and T2DM symptoms. As a result, in the present study, the potential of methanolic and chloroform extracts of E. africanus were investigated ( Figure 8). Unfortunately, none of the extracts exhibited a significant ability to stimulate the β-cell proliferation, instead a decrease in the total cell number was observed, relative to the untreated control after 72 h. This implies that the hypoglycaemia potential of both the methanolic and chloroform extracts of E. africanus is not achieved through the activation of the β-cell proliferation, unlike those reported in the literature by some researchers [89][90][91][92][93].

Pancreatic β-Cell Proliferation
The beta cell apoptosis is a hallmark of both type 1 and types 2 diabetes [84,85]. The islets of the pancreas' beta cells undergo a selective apoptosis, which results in an insulin insufficiency and persistent hyperglycaemia in type 1 diabetes (T1DM). In type 2 diabetes (T2DM), the functional deficiencies and reduced beta cell mass both lead to a beta cell failure [86]. Once T2DM is present, abnormal amounts of metabolic factors cause the beta cell death, which is followed by apoptosis [87]. Strong activators of the β-cell proliferation can restore the pancreatic function through neogenesis [88]. Therefore, showing the therapeutic potential in the relief and or reversal of both the T1DM and T2DM symptoms. As a result, in the present study, the potential of methanolic and chloroform extracts of E. africanus were investigated ( Figure 8). Unfortunately, none of the extracts exhibited a significant ability to stimulate the β-cell proliferation, instead a decrease in the total cell number was observed, relative to the untreated control after 72 h. This implies that the hypoglycaemia potential of both the methanolic and chloroform extracts of E. africanus is not achieved through the activation of the β-cell proliferation, unlike those reported in the literature by some researchers [89][90][91][92][93]. cose homeostasis. Kaempferol glycosides improves the glucose homeostasis by enhancing the pancreatic β-Cell function [94]. The phenolic compounds, such as quercetin and catechin amplified the proliferation inhibition of the pancreatic β-cell [94]. The pancreatic βcell exhibit a sensitivity to oxidative stress, a factor that may lead to the impairment of the β-cell functioning, and thus diabetes.

In Vitro Anti-Inflammatory Activity
Although inflammation is typically thought of as a protective or healing response, many chronic illnesses are marked by persistent inflammation, that leads to tissue dysfunction [95][96][97]. For this rationale, to effectively assess the potential therapeutic importance, the anti-or pro-inflammatory activity of the test samples must be considered within the context of the disease in question, as well as the disease development stage at which intervention would be considered. The mouse macrophage cell line RAW 264.7, which has been extensively studied and is a common model for examining the anti-inflammatory properties of the test extracts, was employed in the present study. Some phenolic compounds, such as kaempferols and dihydrokaempferol, and some of their conjugates, such as kaempferol glucosides, quercetin, and catechin affect the glucose homeostasis. Kaempferol glycosides improves the glucose homeostasis by enhancing the pancreatic β-Cell function [94]. The phenolic compounds, such as quercetin and catechin amplified the proliferation inhibition of the pancreatic β-cell [94]. The pancreatic β-cell exhibit a sensitivity to oxidative stress, a factor that may lead to the impairment of the β-cell functioning, and thus diabetes.

In Vitro Anti-Inflammatory Activity
Although inflammation is typically thought of as a protective or healing response, many chronic illnesses are marked by persistent inflammation, that leads to tissue dysfunction [95][96][97]. For this rationale, to effectively assess the potential therapeutic importance, the anti-or pro-inflammatory activity of the test samples must be considered within the context of the disease in question, as well as the disease development stage at which in-tervention would be considered. The mouse macrophage cell line RAW 264.7, which has been extensively studied and is a common model for examining the anti-inflammatory properties of the test extracts, was employed in the present study.
The anti-inflammatory activity of the methanolic and chloroform extracts of E. africanus was assessed using the RAW 264.7 macrophages and the Griess assay. The cytotoxic effect of the extracts on the RAWs was also determined to accurately establish the potential anti-inflammatory activity. The anti-inflammatory activity is indicated by the decrease in the nitrite concentration, in response to LPS activation of the RAW macrophages, with no effect on the cell viability. The methanolic extract of E. africanus exhibited the antiinflammatory activity from a concentration of 10 µg/mL, while the chloroform extract of E. africanus from 2.5 µg/mL (Figure 9), and interestingly, no significant cytotoxicity was observed from both extracts ( Figure 10). The results from the present study concur with the findings of other researchers, reported in the literature [98]. The inflammatory cytokines, such as IL-1, which cause the beta cell toxicity, are produced, and released because of hyperglycaemia [99]. When the process of glucose intolerance results in overeating, the lipids stored in the non-adipose tissue are ineffectively oxidized; as a result, their products and ceramide promote the nitric oxide generation and induce death in the pancreatic cells [100]. 4,5-Di-O-Caffeoylquinic acid suppresses the inflammatory responses through the TRPV1 activation and was detected in the methanol extract [75]. The anti-inflammatory activity of the methanolic and chloroform extracts of E. africanus was assessed using the RAW 264.7 macrophages and the Griess assay. The cytotoxic effect of the extracts on the RAWs was also determined to accurately establish the potential anti-inflammatory activity. The anti-inflammatory activity is indicated by the decrease in the nitrite concentration, in response to LPS activation of the RAW macrophages, with no effect on the cell viability. The methanolic extract of E. africanus exhibited the anti-inflammatory activity from a concentration of 10 μg/mL, while the chloroform extract of E. africanus from 2.5 μg/mL (Figure 9), and interestingly, no significant cytotoxicity was observed from both extracts ( Figure 10). The results from the present study concur with the findings of other researchers, reported in the literature [98]. The inflammatory cytokines, such as IL-1, which cause the beta cell toxicity, are produced, and released because of hyperglycaemia [99]. When the process of glucose intolerance results in overeating, the lipids stored in the non-adipose tissue are ineffectively oxidized; as a result, their products and ceramide promote the nitric oxide generation and induce death in the pancreatic cells [100]. 4,5-Di-O-Caffeoylquinic acid suppresses the inflammatory responses through the TRPV1 activation and was detected in the methanol extract [75].

In Vitro Macrophage Activation Screening
The outstanding prospects for preventative and therapeutical approaches with minimal adverse side effects are the natural compounds derived from medicinal plants [100][101][102]. A well-established model used for analyzing the anti-inflammatory and macrophage activation capabilities of the test samples is the murine macrophage cell line RAW 264.7. The plant extracts were screened against the RAW 264.7 cells for the macrophage activation potential and the results are represented in Figure 11. The macrophage activation is indicated by an increase in the nitrite concentration, in response to the treatment of the RAW macrophages with no effect on the cell viability, as seen with the LPS treated cells. The plant extracts do not display the potential to activate the macrophages, as compared to the control. However, no toxic effect was recorded at the tested concentrations ( Figure  12). These results revealed that the anti-inflammatory activity demonstrated by the extracts described above, was not achieved through the activation of the macrophages.

In Vitro Macrophage Activation Screening
The outstanding prospects for preventative and therapeutical approaches with minimal adverse side effects are the natural compounds derived from medicinal plants [100][101][102]. A well-established model used for analyzing the anti-inflammatory and macrophage activation capabilities of the test samples is the murine macrophage cell line RAW 264.7. The plant extracts were screened against the RAW 264.7 cells for the macrophage activation potential and the results are represented in Figure 11. The macrophage activation is indicated by an increase in the nitrite concentration, in response to the treatment of the RAW macrophages with no effect on the cell viability, as seen with the LPS treated cells. The plant extracts do not display the potential to activate the macrophages, as compared to the control. However, no toxic effect was recorded at the tested concentrations ( Figure 12). These results revealed that the anti-inflammatory activity demonstrated by the extracts described above, was not achieved through the activation of the macrophages.

Plant Collection and Identification
The fresh leaves of E. africanus were collected behind the Department of Horticultural Science and Textiles building (Coordinate: −33.87507, 18.637135) at Cape Peninsula University of Technology, Bellville campus, South Africa. The plant was authenticated by

Plant Collection and Identification
The fresh leaves of E. africanus were collected behind the Department of Horticultural Science and Textiles building (Coordinate: −33.87507, 18.637135) at Cape Peninsula University of Technology, Bellville campus, South Africa. The plant was authenticated by

Plant Collection and Identification
The fresh leaves of E. africanus were collected behind the Department of Horticultural Science and Textiles building (Coordinate: −33.87507, 18.637135) at Cape Peninsula University of Technology, Bellville campus, South Africa. The plant was authenticated by Prof. Learnmore Kambizi (a botanist at the Department of Horticultural Sciences) and a specimen with the voucher no: 3903 was deposited in their herbarium.

Plant Extraction
The leaves were thoroughly washed with tap and distilled water, to remove impurities, they were air-dried at room temperature for 10 days, and ground with a grinder. One hundred grams (100 g) of the leaf powder was extracted with 1500 mL of 70% methanol and chloroform, in two separate flasks, for 72 h, under constant stirring. The mixture was then filtered using a Whatman no. 1 filter paper and the supernatant was then transferred to a rotary evaporator, where the solvent was evaporated at 48 • C under reduced pressure, to collect the crude extract, which was kept at 4 • C, until further use [21].

Determination of the Total Polyphenol and Flavonol Contents
The Folin-Ciocalteu procedure described by Okafor et al. [103], was explored to determine the total polyphenol content of the extract. The total phenolic content of the extract was measured and represented as mg of the sample gallic acid equivalent (GAE)/g. The total flavonol content was determined using the method reported by Yermakov et al. [104]. The result was represented as milligram catechin equivalents (CE) per gram sample (g).

Estimation of the Antioxidant Capacity
The Trolox equivalent antioxidant capacity (TEAC) of the extract was evaluated using the method outlined by Re et al. [105]. Trolox was employed as a standard drug, and the results were represented as the micromole Trolox equivalent per gram sample (TE)/g. The Benzie and Strain [106] protocol was used for the FRAP analysis. Ascorbic acid was used as the standard antioxidant drug. The potential of the plant extract to scavenge the DPPH radicals was tested using the 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay established by Ngxabi et al. [107]. The results were expressed as M/Trolox equivalent per gram dry weight (mol TE/g). The results were reported as M/Trolox equivalent per gram dry weight (mol TE/g).

Alpha-Glucosidase Inhibition
The alpha-glucosidase inhibition assay was conducted, as described by van de Venter et al. [108]. Briefly, "the plant extracts were sonicated and diluted in an assay buffer (67 mM potassium monobasic anhydrous phosphate pH 6.8) to concentrations of 1000, 500, 250, 125, and 62.5 µg/mL from a final concentration of 100 mg/mL of the dimethyl sulfoxide (DMSO) plant extracts. Thereafter, the extracts were performed in a stepwise manner in 96-well plates, in triplicate. Ten microliters of each extract were added to each well, followed by 70 microliters of the enzyme (Saccharomyces cerevisiae's α-glucosidase), which was then incubated at 37 • C for 10 min. Twenty microliters of the substrate (10 mM p-Nitrophenyl-Dglucopyranoside) was then added, and the mixture was incubated at 37 • C for 20 min and the reaction was terminated by adding 25 µL of Na 2 CO 3 (100 mM). The assay was carried out alongside with the positive control drug, epigallocatechin gallate (EGCG)". Using a BioTek ® PowerWave XS spectrophotometer (Winooski, VT, USA), the absorbance of the mixture was read at 410 nm. The percentage of the α-glucosidase inhibition was estimated as follows in the absence of enzyme and substrate controls: % α-glucosidase inhibition = ((A410 nm of control − A410 nm of test sample))/(A410nm of control) × 100 3.6. Glucose Uptake 3.6.1. Cell Line Maintenance From the Japanese Collection of Research Bioresources Cell Bank, L6 rat myoblast cells were acquired (Osaka, Japan). The routine maintenance of the cells involved maintaining them in 10 cm culture dishes with full media (high glucose DMEM, with 10% FBS and 1% Pen-Strep) and incubating them at 37 • C in a humid environment with 5% CO 2 . The American Type Culture Collection provided human hepatoma-derived C3A hepatocytes (ATCC, Manassas, VA, USA). A complete medium (MEM with 1% NEAA, 10% FBS, and 1% Pen-Strep) was used to maintain the cells in 10 cm culture dishes. The cells were kept at 37 • C in a humidified environment with 5% CO 2 during maintenance.

Glucose Utilization (A) and Uptake (B)
The glucose uptake experiment was carried out with a few modifications, as described by van de Venter et al. [108]. Briefly, "in dimethyl sulfoxide (DMSO), the extracts were reconstituted at a stock concentration of 100 mg/mL. The extracts were sonicated and kept in a freezer until needed. The cells were seeded in 96 well plates (2 × 10 4 cells/well, 100 µL aliquots) and left overnight to attach. Thereafter, the treatments were prepared in a complete medium, added to the cells, and incubated for 24 h. Five microliters (5 µL) of used culture/treatment media were removed and transferred to clean 96-well plates, sealed, and kept at −20 • C until needed (A). The cells were washed with 100 µL of PBS after the leftover media was aspirated. Insulin (1 µg/mL) was employed as a positive control, and 25 µL of the incubation buffer (RPMI-1640 mixed with PBS containing 0.1% BSA to a final glucose concentration of 8 mM) was added to the cells. Following 4 h of incubation, the cells were transferred with 5 µL of the culture media to a fresh 96-well plate (B).
A colorimetric glucose oxidase/peroxidase assay, based on the method described by Trinder [109], was used to determine the changes in the concentration of glucose in the spent culture medium, as a function of the red-colored quinoamine dye complex produced, determined spectrophotometrically at 510 nm. Briefly, "the glucose oxidase assay reagent was freshly prepared, as follows: 3 mM phenol, 0.4 mM 4-aminoantipyrine, 0.25 mM EDTA and 2.5 U/mL horseradish peroxidase in 0.5 M PBS (pH 7.0) with 1 mU/mL glucose oxidase from Aspergillus niger. Two hundred microliters (200 µL) of a glucose oxidase reagent was added to the plates (A and B). The reaction was incubated for 15 min at room temperature. The absorbance was measured at 510 nm using a BioTek ® PowerWave XS spectrophotometer (Winooski, VT, USA). The cell-free wells containing the incubation buffer and complete culture mediums were used as the glucose standards. The glucose uptake and utilization were determined as a function of the concentration of the glucose (mM) remaining and expressed as the difference between the mean of the standard and test extracts.

3-4,5-dimethylthiazol-2,5-diphenyltetrazolium Bromide (MTT) Assay
A MTT assay was performed to ensure that any observed changes in the glucose utilization or uptake were not due to changes in the cell viability. The MTT assay is based on the reduction of a yellow water-soluble tetrazolium salt to an insoluble purple formazan product. The purple-colored formazan crystals are dissolved in dimethyl sulfoxide (DMSO), and the absorbance was measured at 540 nm. The concentration of formazan is then related to the number of healthy viable cells [110,111]. Briefly, the remaining treatment medium from the assay described above was aspirated from all wells and 100 µL of the complete medium, containing 0.5 mg/mL MTT, was added. Thereafter, the cells were incubated for 1 h at 37 • C and a MTT medium was removed and 100 µL DMSO was added to each well. The absorbance was read at 540 nm using a BioTek ® PowerWave XS spectrophotometer (Winooski, VT, USA).

Cell Line Maintenance
The INS-1 pancreatic beta-cells from rat insulinomas were provided by Prof. Rutter from the University of Bristol in England. RPMI-1640 with 10% FBS was used to maintain the cells in 10 cm culture dishes, and they were kept at 37 • C in a humid environment with 5% CO 2 for the whole incubation period.

β-Cell Proliferation Assay
The β-cell proliferation assay was performed, as described by Pringle et al. [88]. Briefly, the cells were seeded in 96-well plates at a density of 2 × 10 4 cells/well in 100 µL aliquots and left overnight to attach. The treatments were prepared in complete medium (RPMI-1640 with 10% FBS) to concentrations of 20, 10, and 5 µg/mL for the methanol extract and 5, 2.5, and 1.25 µg/mL for the chloroform extract, and subsequently, the cells were incubated for 24, 48, and 72 h. Following each incubation period, the culture/treatment medium was gently aspirated, and a 50 µL staining solution [Bis-benzamide H 33342 trihydrochloride (Hoechst) (5 µg/mL) in PBS with Ca 2+ and Mg 2+ ] was added to each well. Subsequently, the plates were incubated at 37 • C for 30 min. Following the incubation period, the fluorescent micrographs were captured with an ImageXpress Micro XLS Widefield Microscope (Molecular Devices) with a 10× Plan Fluor objective and DAPI filter cube. The total number of cells was calculated by analyzing the acquired pictures with the MetaXpress program and the multi-wavelength cell scoring application module.

In Vitro Macrophage Activation
Briefly, the in vitro macrophage activation assay was conducted, following the description of Rampa et al. [112]. Briefly, dimethyl sulfoxide (DMSO) was used to reconstitute the plant extracts to a stock concentration of 100 mg/mL., and the RAW 264.7 cells (Cellonex, South Africa) were seeded at a density of 1 × 10 5 cells per well in 96-well plates containing the RPMI1640 culture media supplemented with 10% FBS (RPMI complete medium), and left overnight to attach. The spent culture media from the previous day was removed, and 50 µL sample aliquots were added and diluted in the RPMI complete medium, to achieve the final concentrations (methanolic extract: 5, 10, 20, and 40 µg/mL; chloroform extract: 1.25, 2.5, 5, and 10 µg/mL). Lipopolysaccharide (LPS) serves as a positive control drug at 500 ng/mL. Subsequently, the cells were incubated for 24 h. The manufacturer's instructions were followed to make the sulfanilamide solution and the NED solution, and 50 µL of the used culture medium was transferred to a fresh 96-well plate to measure the NO production. The used culture media was then mixed with 50 µL of the sulfanilamide solution and allowed to incubate for 10 min in the dark at room temperature. Thereafter, 50 microliters (50 µL) of the NED solution were added to each well, and each well was then incubated for a further 5-10 min, at room temperature and in the dark. Using a BioTek ® PowerWave XS spectrophotometer, the absorbance was read at 540 nm. The concentration of the NO in each sample was estimated using a standard curve constructed using sodium nitrite dissolved in a culture medium. Furthermore, to ascertain that toxicity was not a significant factor, the MTT assay was explored to test the cell viability, as described by van de Venter et al. [108] with minor modifications.

UPLC-ESI-QTOF-MS Analysis of the Methanol and Chloroform Extracts of E. africanus
The LC-MS analysis was conducted using a QA Waters Synapt G2 quadrupole time-offlight mass spectrometer. It was fitted with a Waters ultra-pressure liquid chromatography (UPLC-MS) using Waters msE technology and photodiode array detection. The phenolic method and specification of the instrument were reported by Stander et al. [113], in the negative ion mode with minor modifications. The solvents A and B in the positive ion mode each contained 0.1% formic acid, while the mobile phase in this mode was made up of water and acetonitrile. Following 0.5 min of 100% solvent A, the gradient switched to 100% B for over 0.5 min to 12.5 min. Thereafter, 13 min into the runtime, it then changed to 100% A for the following 2 min in a total run time of 15 min. The flow rate was 0.4 mL/min, the seal wash was 5 min, and the column temperature was maintained at 55 • C. The ionizing electrospray 275 • C desolvation temperature, the 15 V cone voltage, and ESI Pos. Leucine encephalin was injected as a lock mass in the background, and sodium formate was employed for the calibration to obtain the precise mass measurements. The MassLynx software platform supplied with Waters mass spectrometers was used for manually processing each chromatogram.

Identification of the Compounds Using UPLC-ESI-QTOF-MS
The identified metabolites were given preliminary names, based on the accurate mass matches that were automatically searched in the databases, such as Metlin, massBank, NIST, and other libraries, such as PubChem, mass fragmentation patterns of compounds searched in the databases and a number of carbon atoms for the isotope relative abundance. All compounds were identified as unidentified using the accurate mass match, if their accurate mass error (AME) was more than 5 ppm [114]. To identify a particular compound, based on the retention time, the mass fragmentation, and the ionization modes, a few standards of the phenolic compounds were spiked under identical LC/MS conditions (positive and negative ion modes). Considering that it was not possible to obtain all standards and many compounds could be detected using UPLC-MS, the MS and MS2 fragment ions of other compounds, that were similar to those being annotated, were employed. The compound structures were elucidated using the MS-MS analysis of the sample's compounds that were fragmented to match the product ion mass spectra. If isotope abundances were available, the number of carbon atoms in the peak was computed as a final step. The false annotations were minimized by using the predicted number of carbon atoms in the putatively recognized molecule. Different concentrations of the standard mixture of the chemical markers (3.9, 7.8, 15.6, 31.3, 62.5, 125.0, and 250.0 mg/L), were injected for quantification. The linearity of the calibration curve was checked by plotting the peak areas against the series of standard solution concentrations (mg/L) and determining the correlation coefficient using a linear regression model. In all cases, the system was linear when r > 0.99.

Conclusions
In the present study, the phenolic compounds of E. africanus were profiled, and the antidiabetic and anti-inflammatory potentials of the methanolic extract and chloroform leave extracts were investigated in vitro. Both extracts demonstrated a dose-dependent manner for the alpha-glucosidase inhibition and the extracts also demonstrated the potential to increase the glucose utilization at different concentrations in both the L6 cells and C3A cells, and there were no significant cytotoxic effects. However, none of the extracts exhibited a significant ability to stimulate the β-cell proliferation. Moreover, both extracts displayed a strong anti-inflammatory activity with no significant cytotoxic effect, indicating the relative safety in the therapeutic use for humans. The presence of several phytochemical compounds identified in the extracts could be responsible for the biological activities demonstrated in the present study. It is noteworthy that the biological effects displayed by these extracts may perhaps be prompted by the synergistic interaction of the bioactive compounds identified in them. Therefore, further research on the isolation of the pure compounds from these extracts and conducting in vivo tests is encouraged. Overall, the results of this study indicate that E. africanus may be a source of new bioactive compounds to control hyperglycaemia in the treatment of type 2 diabetes mellitus. Thus, further studies, such as the in silico molecular docking of the identified compounds with carbohydrates' digestive enzymes, as well as in vivo studies, are encouraged. Additionally, the targeted isolation of the bioactive compounds responsible for these biological activities is of much significance in studying the biological activities of this plant. Thus, the present findings would be useful for future research directions on the application of traditional medicinal plants in the development of nutraceuticals and pharmaceuticals.